Posts

Sorted by New

Wiki Contributions

Comments

I'm guessing you're aware but Jim Babcock and others have thought a bit about AI containment and wrote about it in Guidelines for AI Containment.

I think Herculano-Houzel would want to mention that humans have 3x (iirc) more neurons in their cerebral cortex than even the elephant species with the biggest brains. Those elephants have more total neurons because their cerebellar cortices have like 200 billion neurons. Humans have more cortical neurons than any animal, including blue whales, because neuron sizes scale differently for different Orders and primates specifically scale well.

Crucially, people have thought human brains were special among primates but she makes the point that it's the other great apes that are special in having smaller brains according to primate brain scaling laws. This is because humans either had a unique incentive to keep up with the costs of scaling or because they had a unique ability to keep up with the costs (due to e.g. cooking).

Having better algorithms that could take advantage of scale fits with her views, I think.

I was under the impression that GPT-4 would be gigantic, according to this quote from this Wired article:

“From talking to OpenAI, GPT-4 will be about 100 trillion parameters,” Feldman says. “That won’t be ready for several years.”

a mathematical formalization of alignment

I can barely see how this is possible if we're talking about alignment to humans, even with a hypothetical formal theory of embedded agency. Do you imagine human values are cleanly represented and extractable, and that we can (potentially very indirectly) reference those values formally? Do you mean something else by "formalization of alignment" that doesn't involve formal descriptions of human minds?

I'm curious if you'd be willing to share a 5 months-later follow up.

1) Great post and great comments.

2) Like a few people have mentioned, using a life force as an explanation isn't necessarily a bad thing. It depends what you have in mind. You could believe in the life force but not be breaking any of the four curiosity stoppers. It would be interesting to know how many people used life force as a curiosity stopper when it was popular. I would guess that most people did use it as a curiosity stopper. Sounds like a good job for those experimental philosophers to show they do more than just polls about intuitions.

3) "You have a little causal diagram in your head that says ["Elan vital!"] -> [hand moves]. But actually you know nothing you didn't know before. You don't know, say, whether your hand will generate heat or absorb heat, unless you have observed the fact already; if not, you won't be able to predict it in advance."

I disagree that you know nothing more than you did before. When I think of a life force I picture different things than, say, electrical force. Maybe your concept of life hasn't substantially changed, but it has been enriched slightly, and the more you enrich a concept the more falsifiable it becomes. I would argue that the more falsifiable a concept is, without being shown to be false, the more useful it is (in general). For instance, if I said meaning was holistic, I think this is somewhat analogous to saying motion in the living is generated by a life force. It loosely constrains other things you can believe about meaning or life.